A charge pump circuit is a basic block for a flash memory and greatly affects the device's program/erase speed. The continuous advancement of integrated circuit (IC) fabrication technology and our relentless pursuit for lower power consumption lead to the development of ICs powered by increasingly lower supply voltages.
On the other hand, in flash memories, program/erase operations of a cell still require relatively high voltages. In such a context, more and more importance is being attached to charge pump circuits in the continuously-developing IC domain. Currently, charge pumps have become a focus of research efforts in the field of flash memory design.
A charge pump, also known as a switched-capacitor voltage converter, is a kind of DC to DC converter that uses so-called “flying” or “pumping” capacitors (rather than inductors or transformers) for energetic charge storage to raise, lower an input voltage or to produce a negative voltage. An internal field effect transistors (FET) switch array of the charge pump controls transfer of electric charge among the capacitors in a certain manner (usually, uses clock signals to control the charge/discharge of the capacitors) so that the input voltage is raised (or reduced) in a certain way to a desired output level.
However, when powered with a supply voltage that is not accurate or varies in a wide range, an existing charge pump tends to produce an inaccurate, unreliable output voltage